You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Our devops agent pool is managed by another team, part of which includes automatic VM version upgrades from the VM images specified at https://github.com/actions/virtual-environments/. Recently we had an issue where the latest VM image included a git cli version upgrade to 2.37.0 which broke our usage of sparse checkout (switching to cone mode by default. See #3606 for more details).
Ideally, we would have been able to catch this issue if we were able to preview new VM image updates coming to our agents several days in advance, before they roll out to the entire VM agent pool. I'm not sure yet what a solution here would look like, whether it's a feature we can ask for from the agent pool team, or whether we would need to build the VM images ourselves and attempt to run a sampling of pipelines on a separate hosted pool.
The text was updated successfully, but these errors were encountered:
ghost
added
the
needs-triage
Workflow: This is a new issue that needs to be triaged to the appropriate team.
label
Jul 11, 2022
kurtzeborn
added
Central-EngSys
This issue is owned by the Engineering System team.
and removed
needs-triage
Workflow: This is a new issue that needs to be triaged to the appropriate team.
labels
Oct 31, 2022
Our devops agent pool is managed by another team, part of which includes automatic VM version upgrades from the VM images specified at https://github.com/actions/virtual-environments/. Recently we had an issue where the latest VM image included a
git
cli version upgrade to2.37.0
which broke our usage of sparse checkout (switching to cone mode by default. See #3606 for more details).Ideally, we would have been able to catch this issue if we were able to preview new VM image updates coming to our agents several days in advance, before they roll out to the entire VM agent pool. I'm not sure yet what a solution here would look like, whether it's a feature we can ask for from the agent pool team, or whether we would need to build the VM images ourselves and attempt to run a sampling of pipelines on a separate hosted pool.
The text was updated successfully, but these errors were encountered: